- Email: [email protected]
Complications of Elbow Arthroscopy in a Community-Based Practice
Journal Pre-proof Complications of Elbow Arthroscopy in a Community Based Practice Jessica Intravia, M.D., Daniel C. Acevedo, M.D., W-L Joanie Chung, MPH, MA, Raffy Mirzayan, M.D. PII:
S0749-8063(19)31098-9
DOI:
https://doi.org/10.1016/j.arthro.2019.11.108
Reference:
YJARS 56687
To appear in:
Arthroscopy: The Journal of Arthroscopic and Related Surgery
Received Date: 2 April 2019 Revised Date:
31 October 2019
Accepted Date: 16 November 2019
Please cite this article as: Intravia J, Acevedo DC, Chung WLJ, Mirzayan R, Complications of Elbow Arthroscopy in a Community Based Practice, Arthroscopy: The Journal of Arthroscopic and Related Surgery (2019), doi: https://doi.org/10.1016/j.arthro.2019.11.108. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier on behalf of the Arthroscopy Association of North America
Complications of Elbow Arthroscopy in a Community Based Practice
Jessica Intravia, M.D.1, Daniel C. Acevedo, M.D.2, W-L Joanie Chung, MPH, MA3, Raffy Mirzayan, M.D.4
1
Department of Orthopaedic Surgery, Philadelphia Hand to Shoulder Center, Jefferson University, Philadelphia PA 2 Department of Orthopaedic Surgery, Kaiser Permanente Southern California, Panorama City, CA 3 Department of Biostatistics, Research and Evaluation, Kaiser Permanente Southern California, Pasadena, CA 4 Department of Orthopaedic Surgery, Kaiser Permanente Southern California, Baldwin Park, CA Please send correspondence to Raffy Mirzayan 1011 Baldwin Park Boulevard Baldwin Park CA 91706
1
Complications of Elbow Arthroscopy in a
2
Community Based Practice
3 4
ABSTRACT
5 6
Purpose: The purpose of this study was to report the complications of elbow arthroscopy in a
7
large community practice with multiple surgeons and to analyze potential risk factors for these
8
complications.
9
Methods: Patient demographic information, surgical variables, surgeon variables, and
10
complications were retrospectively reviewed for all elbow arthroscopies performed
11
within the health network from 2006-2014. Inclusion criteria included patients of any
12
age undergoing a primary and revision elbow arthroscopy, which may have been
13
performed in conjunction with other concomitant procedures. Exclusion criteria
14
included incorrectly coded procedures where arthroscopy was not performed and no
15
postoperative follow-up. Statistical calculations were performed using a Binary Logistic
16
Regression analysis to fit a logistic regression model.
17
Results: 560 consecutive elbow arthroscopies in 528 patients performed between 2006
18
and 2014, by 42 surgeons at 14 facilities were reviewed. 113 procedures were
19
performed in pediatric patients under the age of 18. The average age was 38.6 years
20
(range: 5-88). There were 444 males. The average length of follow up was 375.8 days (2-
21
2739 days). Overall, heterotopic ossification occurred in 14 of 560 (2.5%) cases (all
22
males), and 20 of 560 (3.5%) cases developed transient nerve palsies (8 ulnar, 8 radial, 1
23
median, 3 medial antebrachial cutaneous). There were 3 (0.5%) deep and 11 (2%)
24
superficial infections. There were no vascular injuries, compartment syndrome, deep
25
vein thrombosis, or pulmonary embolism. Elevated blood sugar was a significantly
26
higher risk for infection (OR 4.11, 95% CI: 1.337, 12.645; P=0.0136). Previous elbow
27
surgery (OR 3.57, 95% CI: 1.440, 8.938; P=0.006) and female sex (OR4.05; 95% CI: 1.642,
28
9.970; P=0.002) had a significantly higher risk for nerve injury. Relative to pediatric
29
patients, there was a higher odds in adults for nerve injury, infection, and heterotopic
30
ossification, but none reached significance.
31
Conclusion: Elbow arthroscopy is a safe procedure with low complication rates.
32
Diabetes is a risk factor for infection. Prior surgery and female sex are at higher risk for
33
nerve injury.
34
Level of Evidence: Case Series, Level 4
35
Key Terms: elbow arthroscopy, complications, nerve injury
36
37 38
INTRODUCTION Elbow arthroscopy has been described to treat a variety of conditions including
39
arthritis, symptomatic loose body removal, elbow contracture, valgus extension
40
overload, impingement, osteochondritis dessicans, lateral epicondylitis, septic joint, and
41
to aid in the diagnosis of elbow pain. As the surgical indications for elbow arthroscopy
42
expand, the procedures also have become increasingly complex.1 With increasing
43
complexity and popularity of elbow arthroscopy, concern remains over the safety of the
44
procedure, specifically in regards to nerve injury. Previous studies have reported nerve
45
injury rates from 0.5 - 10% with the ulnar nerve most frequently injured.2-4
46
Several surgeons have published their personal experience and outcomes with
47
elbow arthroscopy; however, few papers report on a large series of consecutive
48
patients, and in a community practice. (Table 1) Kelly et al report his institution’s
49
experience with 473 consecutive elbow arthroscopies performed by twelve surgeons
50
from 1980-1998; however, the majority of these procedures (89%) were performed by
51
one of two senior surgeons.5 Nelson et al shared the results of 417 consecutive elbow
52
arthroscopies performed over a thirteen year period by three orthopedic surgeons who
53
have completed specialized subspecialty training in shoulder and elbow surgery.1 While
54
these studies provide important insight into the complications associated with elbow
55
arthroscopy, they may fail to capture the true rate of complications in the general
56
community observed with elbow arthroscopy. The purpose of this study was to report
57
the complications of elbow arthroscopy in a large community practice with multiple
58
surgeons and to analyze potential risk factors for these complications. Our hypothesis
59
was that the rate of complications, specifically in regards to nerve injury, infections and
60
heterotopic ossification, would be similar to previously published reports, and that
61
diabetes would be a risk factor for infection.
62
METHODS
63 64
After obtaining institutional review board approval, a single reviewer (JI) not
65
involved with the original procedures reviewed all elbow arthroscopies performed
66
between January 1, 2006 and December 31, 2014 in a multispecialty, integrated health
67
care system by 42 surgeons at 14 medical centers. The institutional review board
68
incorporated all 14 hospitals, and given the retrospective study design, individual
69
surgeon or patient approval was not required. Our institution does not code surgical
70
procedures by CPT codes, rather by inter-facility codes. Elbow arthroscopies were
71
identified by using the following inter-facility codes: “elbow arthroscopy” and “elbow
72
debridement arthroscopic”.
73
Inclusion criteria included patients of any age undergoing a primary and revision
74
elbow arthroscopy, which may have been performed in conjunction with other
75
concomitant procedures. Exclusion criteria included incorrectly coded procedures
76
where arthroscopy was not performed and no postoperative follow-up. All operative
77
notes, orthopedic outpatient and inpatient visit documentation, and urgent care and
78
emergency room visits were reviewed. All patients were part of an integrated single
79
payer health system with a single electronic medical record. One patient had no further
80
encounters after his initial surgery and was removed from the analysis of postoperative
81
complications. Release from medical care was defined as the length of time from
82
surgery to the last visit with an orthopedic provider.
83
Patient demographic information was recorded, as well as smoking status, body
84
mass index, medical comorbidities, hand dominance, surgical history of operative
85
extremity, preoperative, intraoperative and final postoperative elbow range of motion
86
of the operative extremity, and need for additional procedures on the operative
87
extremity. Due to limitations in charting and variability amongst providers, any patient
88
with a listing of diabetes, prediabetes, or high blood sugar on his medical record
89
problem list was categorized under a single “elevated blood sugar” category.
90
Surgical variables such as preoperative diagnosis, detailed nature of procedures
91
performed, tourniquet time, patient position, portals used, need for concomitant ulnar
92
nerve procedure, need for unplanned arthrotomy, use of local anesthesia, and use of
93
intraarticular steroid were recorded. Complications such as superficial infection, deep
94
infection, nerve injury, vascular injury, compartment syndrome, heterotopic ossification,
95
future surgeries, deep vein thrombosis/ pulmonary embolism, and re-hospitalization
96
were also documented. Deep infection was defined as any postoperative infection
97
requiring hospitalization, intravenous antibiotics and/or surgical debridement.
98
Superficial infection was defined as any patient prescribed post-operative oral
99
antibiotics which did not require hospitalization or surgical irrigation and debridement.
100
Nerve damage was further stratified by nerve injured: ulnar, radial, median or medial
101
antebrachial cutaneous nerve. Any patient with concomitant open ulnar nerve
102
procedure was excluded from the analysis of ulnar nerve damage, as it would be unclear
103
if the nerve deficit were related to the arthroscopy or ulnar nerve procedure.
104
Surgeon characteristics were gathered including fellowship training, years in
105
practice, and number of cases performed during study period. For the purpose of this
106
study, pediatric patients were defined as age eighteen and younger at time of the
107
surgery. One of the authors who did not perform any of the surgical procedures
108
compiled all the data through a retrospective chart review.
109
To aid in the analysis of risk factors and the perceived risk of the procedure, the
110
complexity scale designed by Nelson et al was modified (Table 2).1 This is not a validated
111
scale. Nelson et al tracked the factors contributing to the complexity rating (ranging
112
from 1 to 9) including type of procedure (scored as 1 to 5), tourniquet time (scored as 0-
113
2), and number of portals used (scored as 0-2). The procedure was then categorized as
114
low, medium, or high complexity based on its numeric score. In the modified complexity
115
scale, the type of procedures scored was expanded to include a wider variety of
116
procedures. Nelson et al allotted a bonus point for release of posterior band of medial
117
collateral ligament or medial epicondylectomy. In the modified scale, the bonus point
118
was awarded for any concomitant ulnar nerve procedure.
119
Statistical Analysis
120
Statistical calculations were performed using a Binary Logistic Regression
121
analysis to fit a logistic regression model to investigate the relationship between
122
discrete outcomes with binary levels and a set of categorical predictors. Results were
123
determined to be statistically significant with a p–value of less than 0.05.
124 125 126
RESULTS Five hundred and seventy-three procedures were identified. Of those, 560
127
elbow arthroscopies in 528 patients met our inclusion criteria. Twelve procedures were
128
excluded due to incorrect inter-facility code. One procedure was excluded due to lack of
129
followup. Of those procedures, 447 (79.8%) were performed in adults and 113 (20.2%)
130
in pediatric patients. The mean age at the time of surgery was 38.6 years (range 5 – 88
131
years). The mean age in adults was 44.5 years (19-88 years) and the mean age in
132
pediatric patients was 15.2 years (5-18 years). 79.1% of the procedures were performed
133
in male patients. Five hundred and twelve patients (91.4%) did not use tobacco at the
134
time of the procedure. Fifty procedures (8.9%) were performed in patients who had
135
diabetes and twenty patients (3.6%) had prediabetes at the time of their procedure.
136
Thirteen patients (2.3%) had a diagnosis of rheumatoid arthritis and two patients (0.4%)
137
had a diagnosis of juvenile rheumatoid arthritis at the time of their procedure. One
138
hundred and nine patients (19.5%) had had a previous elbow surgery on the operative
139
extremity. Twelve procedures (2.1%) received and intra-articular steroid injection. Pre-
140
procedure diagnosis included elbow arthritis, symptomatic loose bodies, elbow
141
contracture, valgus extension overload, impingement, instability, osteochondritis
142
dessicans, lateral epicondylitis, septic joint, and elbow pain. The complexity of cases is
143
summarized in Figure 1.
144
The average length of follow-up was 376 days (range 2- 2739 days). Five patients
145
had less than a week follow-up. An additional forty-three patients had less than two
146
weeks follow-up. Fifty-nine elbows (10.5%) required a future surgery on the operative
147
elbow (Table 3). Seventy-one patients (12.7%) had a concomitant ulnar nerve
148
procedure. These were excluded from the analysis of ulnar nerve complications
149
postoperatively.. 367 (65.5%) procedures were performed in the prone position, 178
150
(31.8%) in lateral, and 10 (1.8%) in supine and 1 (0.2%) in beach chair.
151
Of the 42 board certified orthopaedic surgeons, 33 (78.6%) of surgeons were
152
fellowship trained. The fellowships were in sports medicine (38.0%), hand/wrist
153
(31.0%), shoulder and elbow (7.1%) or upper extremity (2.4%). Of the operating
154
surgeons, 33 (78.6%) performed less than 10 elbow arthroscopies in the study sample, 5
155
(11.9%) performed 11-50 elbow arthroscopies, 3 (7.1%) performed 51-100 and 1 (2.4%)
156
performed greater than 100 elbow arthroscopies in the study period. Ninety-two
157
(16.4%) procedures were performed during the surgeon’s first 5 years in practice. Two
158
hundred and forty-one (43.0%) procedures were performed between the surgeon’s 6
159
and 10 years in practice. Ninety-five (17.0%) procedures were performed between the
160
surgeon’s 11 and 15 years in practice. Forty-one (7.7%) procedures were performed
161
between the surgeon’s 16 and 20 years in practice. Sixteen (2.9%) procedures were
162
performed after the surgeon had been in practice more than 20 years. Surgeon based
163
data was unavailable for seventy-five procedures.
164 165
Complications
166
Nerve Injuries
167
The most frequent reported complication was transient neurologic
168
complications. Of the 560 procedures, there were twenty transient nerve palsies (3.5%).
169
Nerve injuries included: 8 (1.4%) ulnar (7 sensory, 1 mixed motor/sensory), 8 (1.4%)
170
radial/posterior interosseous nerve (6 sensory, 1 motor to posterior interosseous nerve,
171
1 mixed motor/sensory radial nerve), 3 (0.5%) medial antebrachial cutaneous nerve, and
172
1 (0.2%) median nerve (sensory). Table 4 summarizes the details of each case. Previous
173
elbow surgery (OR 3.23, P=0.004) and female sex (OR2.96; P=0.008) had a significantly
174
higher risk of nerve injury (Table 5).
175
176 177
Infection There were 3 (0.5%) deep infections and eleven (2%) superficial infections. One
178
deep infection was culture negative, one had positive mycobacterium avium cultures
179
which eventually required resection arthroplasty. A third late infection (6 month post
180
op) was culture positive for streptococcus viridans. Table 6 summarizes all the variables
181
that may play a role in causing an infection. The only variable that reached significance
182
for infection was elevated blood sugars. Patients with elevated blood sugars had a 4.11
183
odds (95% CI: 1.337, 12.645; P=0.0136) of developing a post-operative infection. There
184
were no infections in patients who were smokers or those who had a steroid injection at
185
completion of case.
186
187 188 189
Heterotopic Ossification
190
with nine (1.6%) requiring additional surgeries for resection. All cases of HO occurred in
191
male patients, but no other significant risk factors were identified (Table 7), Figure 2.
192
Adult vs Pediatric
Fourteen (2.5%) elbows developed postoperative heterotopic ossification (HO)
Relative to pediatric patients, there no statistically significant increased risk
193 194
adults for nerve injury (OR 1.99; P=0.27), infection (OR 3.286; P=0.24), and heterotopic
195
ossification (OR 1.40; P=0.66). There was no difference in re-operation rate (OR 0.92,
196
P=0.8).
197 198 199
Other Complications There were no cases of vascular injury, compartment syndrome, deep vein
200
thrombosis, or pulmonary embolism. Three patients required re-hospitalization for a
201
deep infection post operatively.
DISCUSSION
202 203
In this large case series with forty-two different operating surgeons, 3.5% of
204
procedures developed a transient nerve palsy. Previous elbow surgery (OR 3.57,
205
P=0.006) and female sex (OR4.05; P=0.002) were significant risk factors for nerve injury.
206
In addition, 2.5% of cases developed postoperative heterotopic ossification, all of whom
207
were male. No other significant risk factors were identified. Relative to pediatric
208
patients, there was a higher odds in adults for nerve injury, infection, and heterotopic
209
ossification, but none reached significance. We also found that patient factors (BMI,
210
age), surgical factors (patient position, tourniquet time, complexity) and surgeon factors
211
(fellowship training, case volume, years of practice) did not have a significance in
212
neurologic complications.
213
Reported nerve injury rates range from 0.5% -10.4% (Table 7). 1-8 Recent large
214
case series report neurological complications of approximately 2%. These are lower than
215
our reported 3.5% nerve injury rate. The reason behind this discrepancy is unclear. One
216
potential explanation is our inclusion of concomitant arthroscopic and open procedures.
217
We chose to include all surgeries with an arthroscopic component as we felt this was
218
more indicative of everyday community practice.10 ,11 Many prior papers have focused
219
on arthroscopic surgeries alone. Four hundred and twenty one (75%) of our procedures
220
were exclusively arthroscopic.
221
Heterotopic ossification may be an underreported complication of elbow
222
arthroscopy. In this series, 2.5% of cases developed postoperative heterotopic
223
ossification with 1.6% of cases requiring repeat surgical procedures as a result of the
224
heterotopic ossification. All cases occurred in male patients, but no other significant risk
225
factors were identified. Heterotopic ossification prophylaxis was not routinely utilized.
226
Similar to our results, Nelson et al reported 6 cases (1.4%) of heterotopic ossification
227
requiring future surgeries. Nelson notes that postoperative XR were not regularly
228
obtained and that the six-week follow-up of his study may have led to the under
229
reporting of this complication. Likewise, not all surgeons in our study obtained routine
230
postoperative XR. Future studies are warranted to further examine the risk of post-
231
operative heterotopic ossification and possible prevention strategies.
232
Post- operative infection is an important complication of elbow arthroscopy. We
233
found a 2% superficial infection and 0.5% deep infection which was similar to published
234
results.12 Nelson et al published a substantially higher infection rate (6.7% superficial
235
infection and 2.2% deep infection). Previous studies have documented the use of intra-
236
articular steroid injection as a risk factor for infection. In this study, steroid injection was
237
not commonly performed (2.1% of procedures). No infections were seen in patients
238
who received a steroid injection. Overall, elevated blood sugar had a significantly higher
239
risk for infection (OR 4.11, P=0.0136). Smoking status, body mass index greater than 40,
240
tourniquet time and previous elbow surgery did not significantly increase the risk of
241
infection. There were no cases of vascular injury, compartment syndrome, deep vein
242
thrombosis, or pulmonary embolism.
243
Our study is unique in that it contains a large pediatric cohort. Few previous
244
large-scale studies have been published on elbow arthroscopy in the pediatric
245
population.13 Vavken et al published on the experience of the first 50 elbow
246
arthroscopies performed in a pediatric orthopedic department in Switzerland. They had
247
3 transient nerve palsies (6%) in this series.14 Micheli et al also published on their
248
experience with 47 elbow arthroscopies in the pediatric and adolescent population.15
249
They had no nerve palsies in their series. We found two nerve palsies out of the 114
250
pediatric elbow arthroscopies studied (1.8%) – of these palsies, one was the ulnar nerve
251
and one was the radial nerve. Relative to pediatric patients, there was a higher risk in
252
adults for nerve injury (OR 1.99; P=0.27), infection (OR 3.286; P=0.24), and heterotopic
253
ossification (OR 1.40; P=0.66). However, none of these trends reached statistical
254
significance. There was no difference in re-operation rate (OR 0.92, P=0.8).
255
Strengths
256 257
The strengths of this study includes the large number of cases and contributing surgeons which help provide a realistic analysis of post-operative complications that can
258
be applicable to many community orthopedic surgeons. The inclusion of surgeries with
259
both an arthroscopic and concomitant open component increases the applicability of
260
these results and aides with preoperative patient counseling. This study examined
261
specific risk factors for postoperative complications and determined odds ratios for each
262
risk factor identified. This allows for a more in-depth analysis than database studies
263
allow.
264
Limitations
265
The weakness of this study is that it is retrospective. The integrated nature of the health
266
system allows for the vast majority of patients to be captured in a single electronic record.
267
However, a patient could theoretically seek care outside of the health care system for a
268
postoperative complication. This would be unlikely given the financial implications of an
269
outside facility. In addition, the small number of complications made statistical analysis
270
difficult and the study was not powered to reflect differences between the adult and
271
pediatric populations. The patients with concomitant ulnar nerve releases were excluded
272
from the analysis of post-operative ulnar nerve complications. Given that these patients
273
complained of pre-operative ulnar nerve symptoms, the authors were unable to tell from
274
the medical record if post-operative symptoms were a result of the underlying nerve
275
health or a result of a new injury. This may lead to an under reporting of ulnar nerve
276
injuries. In addition, the complexity scale we used was not validated. In our series, 48
277
patients had less than 2-week follow up which may have an inadequate length of follow-
278
up to properly analyze certain complications. Additional limitations include no defined
279
minimum follow up period, postoperative elbow stiffness and range of motion were not
280
analyzed complications in this study.
281 282 283 284
Conclusions Elbow arthroscopy is a safe procedure with low complication rates. Diabetes is a risk factor for infection. Prior surgery and female sex are at higher risk for nerve injury.
285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
References 1. Nelson GN, Wu T, Galatz LM, Yamaguchi K, Keener JD. Elbow arthroscopy: early complications and associated risk factors. JSES 2014(23):273-278. 2. Andrews JR, Carson WG. Arthroscopy of the elbow. Arthroscopy 1985(1):97-107. 3. O’Driscoll SW, Morrey BF. Arthroscopy of the elbow: Diagnostic and therapeutic benefits and hazards. J Bone Joint Surg Am. 1992(74): 84-94. 4. Reddy AS, Kvitne RS, Yocum LA, Elattrache NS, Glousman RE, Jobe FW. Arthroscopy of the elbow: a long-term clinical review. Arthroscopy 2000 (16): 588-94. 5. Kelly EW, Morrey BF, O’Driscoll SW. Complications of Elbow Arthroscopy. J Bone Joint Surg Am. 2001 (83):25-34. 6. Jinnah AH et al. Peripheral Nerve Injury After Elbow Arthroscopy: An Analysis of Risk Factors. Arthroscopy. 2018; 34(5): 1447-1452. 7. Marti D, Spross C, Jost B. The first 100 elbow arthroscopies of one surgeon: analysis of complications. JSES 2013 (22): 567-573. 8. Elfeddali R, Schreuder MH, Eygendaal D. Arthroscopic elbow surgery, is it safe? JSES 2013 (22): 647-652. 9. Schneider T, Hoffstetter I, Fink B, Jerosch J. Long-term results of elbow arthroscopy in 67 patients. Acta Orthop Belg. 1994 (60): 378-83. 10. Hilgersom NFJ, van Deurzen DFP, Gerritsma CLE, van der Heide HJL, Malessy MJA, Eygendaal D, van den Bekerom MPJ. Nerve injuries do occur in elbow arthroscopy. Knee Surg Sports Traumatol Arthrosc 2018: 26(1): 318-324. 11. Desai MJ, Mithani SK, Lodha SJ, Richard MJ, Leversedge FJ, Ruch DS. Major Peripheral Nerve Injuries After Elbow Arthroscopy. Arthroscopy. 2016; 32(6):999-1002. 12. Campl CL, Cancienne JM, Degen RM, Dines JS, Altchek DW, Werner BC. “Factors that Increase the Risk of Infection After Elbow Arthroscopy: Analysis of Patient Demographics, Medical Comorbidities, and Steroid Injections in 2,704 Medicare Patients. Arthroscopy. 2017: 33(6); 1175-1179. 13. Andelman SM, Meier KM, Walsh AL, Hausman MR. Pediatric elbow arthroscopy: indications and safety. J Shoulder Elbow Surg. 2017: 26(10):1862-1866. 14. Vavken, Muller, Camathias. First 50 Pediatric and Adolescent Elbow Arthroscopies: Analysis of Indications and Complications. Journal of Pediatric Orthopaedics. 2016: 36(4): 400-404. 15. Micheli, Luke, Mintzer, Waters. Elbow Arthroscopy in the pediatric and adolescent population. Arthroscopy. 2001: 17(7):694-9.
323 324
Table 1: Published Nerve Injuries after Elbow Arthroscopy Paper
Year
Number
Total nerve
Ulnar
Radial
MABC
Median
Other
Jinnah et al
Nelson et al Marti et al
injuries (%)
2018
253
15 (5.9%)
11
0
1
1
2
2014
417
7 (1.7%)
5
0
0
1
1
2013
100
2 (2%)
1
0
0
1
0
2013
200
4 (2%)
3
1
0
0
0
2001
473
12 (2.5%)
5 (1 transected)
5
1
1
0
2000
172
1 (0.5%)
1 (transected)
0
0
0
0
8
1994
67
7 (10.4%)
2
4
0
1
0
3
1992
71
3 (4%)
0
3
0
0
0
1985
12
1 (8.3%)
0
0
1
0
0
1
6
Elfeddali et al. Kelly et al
of Cases 6
7
5
Reddy et al
4
Schneider et al O’Driscoll et al Andrews et al
325
2
326
Table 2: Modified Elbow Arthroscopy1 Complexity Scale Modified Elbow Arthroscopy Complexity Scale Procedure Limited debridement (1 compartment), diagnostic arthroscopy, irrigation and debridement Extensive debridement (2+ compartments), loose body removal, plica excision, osteophyte/mass excision, olecranon bursectomy Capsular release, microfracture, radial head resection, ligament repair, lateral epicondylar release Capsular resection with resection of bone, osteochondral allograft, osteochondral lesion fixation, Outerbridge-Kashiwagi, radial head replacement/ ORIF, UCL reconstruction Ulnar Nerve Procedure Tourniquet Time ≤ 60 minutes 60-90 minutes > 90 minutes Portals ≤2 3-4 >4 Complexity Category Low Moderate High
Points 1 2 3 4
+1 0 1 2 0 1 2 0-3 4-5 >5
327 328
Table 3: Repeat Operative Procedures after Elbow Arthroscopy Repeat Operative Procedures Ulnar nerve decompression
11
Open irrigation and debridement, loose body removal
10
Repeat arthroscopic debridement
9
Heterotopic ossification resection
8
Total elbow arthroplasty
5
Open contracture release
5
Radial nerve decompression
4
Radial head replacement
3
Open elbow exploration
2
Medial collateral ligament reconstruction
1
Osteocapsular arthroplasty
1
Scar excision
1
Osteochondral defect surgery
1
Medial epicondylar release
1
Lateral epicondylar release
1
329 330
Table 4: Nerve Injuries after Elbow Arthroscopy Patient ID
Nerve Injured
Procedure Performed
61F
Radial nerve sensory palsy improved at 4 months
Capsular release and osteophyte removal
43F
Radial nerve sensory palsy improved at 4 weeks
Arthroscopic extensor carpi radialis brevis debridement
Low
15M
Radial nerve sensory palsy resolved at 3 months
Anterior capsular release
Moderate
33M
Radial nerve sensory palsy resolved by 6 weeks
Debridement of plica, ulnar nerve decompression
Moderate
50F
Radial nerve sensory palsy resolved by 7 months
Arthroscopic debridement, open capitellar resurfacing using tendon allograft, ulnar nerve decompression
Moderate
36M workers compensation
Radial nerve sensory palsy improved at 11 months
Debridement, capsulectomy, synovectomy, osteophyte resection, ulnar nerve transposition
High
57F
Posterior interosseous nerve motor palsy improving at 6 months
Debridement, capsulectomy, partial synovectomy
Moderate
35F with rheumatoid arthritis and diabetes mellitus
Radial nerve motor and sensory palsy with post operative septic elbow, nerve palsy resolved by 2 years
Debridement followed by arthroscopic irrigation and debridement
Moderate
20F
Median nerve sensory palsy resolved by 2 months
Posteromedial elbow debridement, microfracture
Moderate
Modified Complexity Level Low
331 332
45M with HTN, hyperlipidemia and diabetes
Ulnar nerve sensory palsy resolved by 2.5 months
Debridement of plica
Low
53F
Ulnar nerve sensory palsy continued at 2 weeks
Debridement, capsular division, partial synovectomy
Moderate
17F
Ulnar nerve sensory palsy resolved by two years
Arthroscopic posterior osteophyte excision and release with open anterior release
Moderate
31M
Ulnar nerve sensory palsy resolved by 2 months
Debridement and medial collateral ligament reconstruction using allograft tissue
Moderate
60M with HTN, CAD, Ehlers Danlos, emphysema
Ulnar nerve sensory palsy resolved by 6 weeks
Outerbridge-Kashiwagi procedure
Moderate
28M
Ulnar nerve sensory palsy resolved by 6 weeks
Osteocapsular arthroplasty
High
52M
Ulnar nerve sensory palsy, improved 4m post op
Synovectomy
Moderate
38M
Ulnar nerve sensory and motor palsy, underwent ulnar nerve exploration and release at 2 years with intact nerve
Debridement, osteophyte resection, capsular division, partial synovectomy, injection of amniotic cells into joint space
High
67F with hypertension, hyperlipidemia and obesity
Medial antebrachial cutaneous nerve sensory palsy improving at 6 weeks
Endoscopic carpal tunnel release, open anterior ulnar transposition, elbow arthroscopy, removal of loose body and debridement
Moderate
24F with history of alcohol abuse and anxiety
Medial antebrachial cutaneous nerve sensory palsy improving at 3 months
Elbow arthroscopy, debridement, osteophyte resection, capsular division, synovectomy, loose body removal, ulnar nerve transposition
High
64M
Medial antebrachial cutaneous nerve sensory palsy improving at 2 weeks, resolved by 2 years
Elbow arthroscopy and extensive debridement, removal of loose body
High
333
Table 5 summarizes the odds of each variable for sustaining a nerve injury Odds ratio
95% CI
P-value
Age >18 years
2.328
0.532
10.183
0.2616
BMI >40
3.625
0.775
16.965
0.1019
Tourniquet Time >60 minutes
0.983
0.350
2.762
0.9735
Previous Elbow Surgery
3.567
1.440
8.839
0.0060
Moderate Complexity (4-5)
2.921
0.812
10.513
0.1009
High Complexity >5
1.985
0.466
8.453
0.3534
Supine vs lateral positioning
2.714
0.301
24.494
0.3737
Prone vs lateral positioning
0.826
0.319
2.135
0.6928
Supine vs prone positioning
3.287
0.385
28.067
0.2768
Female sex
4.047
1.642
9.970
0.0024
More than 5 years post training
2.397
0.312
18.397
0.4004
Smoker (yes)
0.552
0.072
4.216
0.5668
Fellowship-trained (Yes)
0.722
0.092
5.679
0.7567
334 335
Table 6 summarizes the odds of each variable for developing an infection Odds ratio
95% CI
P-value
Age >65 years
1.090
0.139
8.566
0.9349
Age >18 years
3.355
0.434
25.918
0.2459
BMI >40
2.385
0.295
19.289
0.4152
NA
NA
NA
0.9739
Elevated blood sugar
4.112
1.337
12.645
0.0136
Tourniquet Time > 60 minutes
0.527
0.169
1.639
0.2686
History of previous elbow surgery
2.340
0.768
7.128
0.1346
Moderate Complexity (4-5)
1.061
0.295
3.820
0.9273
High Complexity (>5)
1.173
0.289
4.788
0.8209
More than 4 portals
0.488
0.063
3.782
0.4919
Male
1.601
0.353
7.254
0.5416
Smoker (yes)
More than 5 years in practice
0.801
0.173
3.701
0.7761 0.5158
Fellowship-trained
336 337
Table 7 summarizes the odds of each variable for developing heterotopic ossification Odds ratio
95% CI
P-value
Age >18 years
1.403
0.307
6.417
0.6627
BMI > 40
2.583
0.317
21.020
0.3749
Tourniquet Time > 90minutes
1.910
0.529
6.903
0.3234
History of previous elbow surgery
1.237
0.335
4.573
0.7499
Moderate Complexity (4-5)
2.157
0.430
10.818
0.3500
High Complexity >5
2.996
0.573
15.677
0.1938
Use of small arthroscope
1.813
0.226
14.517
0.5753
Smoker (yes)
0.906
0.115
7.123
0.9251
Fellowship-trained
0.5312
338 339 340
Figure 1: Number of cases performed in each complexity category as detailed in the Modified Elbow Complexity Scale (Table 2)
341 342 343 344
Figure 2: Representative case of 40 year old male with left elbow contracture (A) Patient underwent arthroscopic osteocapsular arthroplasty. (B) Initial post-operative radiographs. (C) Patient subsequently developed heterotopic ossification requiring open heterotopic ossification excision.
S0749-8063(19)31098-9
DOI:
https://doi.org/10.1016/j.arthro.2019.11.108
Reference:
YJARS 56687
To appear in:
Arthroscopy: The Journal of Arthroscopic and Related Surgery
Received Date: 2 April 2019 Revised Date:
31 October 2019
Accepted Date: 16 November 2019
Please cite this article as: Intravia J, Acevedo DC, Chung WLJ, Mirzayan R, Complications of Elbow Arthroscopy in a Community Based Practice, Arthroscopy: The Journal of Arthroscopic and Related Surgery (2019), doi: https://doi.org/10.1016/j.arthro.2019.11.108. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier on behalf of the Arthroscopy Association of North America
Complications of Elbow Arthroscopy in a Community Based Practice
Jessica Intravia, M.D.1, Daniel C. Acevedo, M.D.2, W-L Joanie Chung, MPH, MA3, Raffy Mirzayan, M.D.4
1
Department of Orthopaedic Surgery, Philadelphia Hand to Shoulder Center, Jefferson University, Philadelphia PA 2 Department of Orthopaedic Surgery, Kaiser Permanente Southern California, Panorama City, CA 3 Department of Biostatistics, Research and Evaluation, Kaiser Permanente Southern California, Pasadena, CA 4 Department of Orthopaedic Surgery, Kaiser Permanente Southern California, Baldwin Park, CA Please send correspondence to Raffy Mirzayan 1011 Baldwin Park Boulevard Baldwin Park CA 91706
1
Complications of Elbow Arthroscopy in a
2
Community Based Practice
3 4
ABSTRACT
5 6
Purpose: The purpose of this study was to report the complications of elbow arthroscopy in a
7
large community practice with multiple surgeons and to analyze potential risk factors for these
8
complications.
9
Methods: Patient demographic information, surgical variables, surgeon variables, and
10
complications were retrospectively reviewed for all elbow arthroscopies performed
11
within the health network from 2006-2014. Inclusion criteria included patients of any
12
age undergoing a primary and revision elbow arthroscopy, which may have been
13
performed in conjunction with other concomitant procedures. Exclusion criteria
14
included incorrectly coded procedures where arthroscopy was not performed and no
15
postoperative follow-up. Statistical calculations were performed using a Binary Logistic
16
Regression analysis to fit a logistic regression model.
17
Results: 560 consecutive elbow arthroscopies in 528 patients performed between 2006
18
and 2014, by 42 surgeons at 14 facilities were reviewed. 113 procedures were
19
performed in pediatric patients under the age of 18. The average age was 38.6 years
20
(range: 5-88). There were 444 males. The average length of follow up was 375.8 days (2-
21
2739 days). Overall, heterotopic ossification occurred in 14 of 560 (2.5%) cases (all
22
males), and 20 of 560 (3.5%) cases developed transient nerve palsies (8 ulnar, 8 radial, 1
23
median, 3 medial antebrachial cutaneous). There were 3 (0.5%) deep and 11 (2%)
24
superficial infections. There were no vascular injuries, compartment syndrome, deep
25
vein thrombosis, or pulmonary embolism. Elevated blood sugar was a significantly
26
higher risk for infection (OR 4.11, 95% CI: 1.337, 12.645; P=0.0136). Previous elbow
27
surgery (OR 3.57, 95% CI: 1.440, 8.938; P=0.006) and female sex (OR4.05; 95% CI: 1.642,
28
9.970; P=0.002) had a significantly higher risk for nerve injury. Relative to pediatric
29
patients, there was a higher odds in adults for nerve injury, infection, and heterotopic
30
ossification, but none reached significance.
31
Conclusion: Elbow arthroscopy is a safe procedure with low complication rates.
32
Diabetes is a risk factor for infection. Prior surgery and female sex are at higher risk for
33
nerve injury.
34
Level of Evidence: Case Series, Level 4
35
Key Terms: elbow arthroscopy, complications, nerve injury
36
37 38
INTRODUCTION Elbow arthroscopy has been described to treat a variety of conditions including
39
arthritis, symptomatic loose body removal, elbow contracture, valgus extension
40
overload, impingement, osteochondritis dessicans, lateral epicondylitis, septic joint, and
41
to aid in the diagnosis of elbow pain. As the surgical indications for elbow arthroscopy
42
expand, the procedures also have become increasingly complex.1 With increasing
43
complexity and popularity of elbow arthroscopy, concern remains over the safety of the
44
procedure, specifically in regards to nerve injury. Previous studies have reported nerve
45
injury rates from 0.5 - 10% with the ulnar nerve most frequently injured.2-4
46
Several surgeons have published their personal experience and outcomes with
47
elbow arthroscopy; however, few papers report on a large series of consecutive
48
patients, and in a community practice. (Table 1) Kelly et al report his institution’s
49
experience with 473 consecutive elbow arthroscopies performed by twelve surgeons
50
from 1980-1998; however, the majority of these procedures (89%) were performed by
51
one of two senior surgeons.5 Nelson et al shared the results of 417 consecutive elbow
52
arthroscopies performed over a thirteen year period by three orthopedic surgeons who
53
have completed specialized subspecialty training in shoulder and elbow surgery.1 While
54
these studies provide important insight into the complications associated with elbow
55
arthroscopy, they may fail to capture the true rate of complications in the general
56
community observed with elbow arthroscopy. The purpose of this study was to report
57
the complications of elbow arthroscopy in a large community practice with multiple
58
surgeons and to analyze potential risk factors for these complications. Our hypothesis
59
was that the rate of complications, specifically in regards to nerve injury, infections and
60
heterotopic ossification, would be similar to previously published reports, and that
61
diabetes would be a risk factor for infection.
62
METHODS
63 64
After obtaining institutional review board approval, a single reviewer (JI) not
65
involved with the original procedures reviewed all elbow arthroscopies performed
66
between January 1, 2006 and December 31, 2014 in a multispecialty, integrated health
67
care system by 42 surgeons at 14 medical centers. The institutional review board
68
incorporated all 14 hospitals, and given the retrospective study design, individual
69
surgeon or patient approval was not required. Our institution does not code surgical
70
procedures by CPT codes, rather by inter-facility codes. Elbow arthroscopies were
71
identified by using the following inter-facility codes: “elbow arthroscopy” and “elbow
72
debridement arthroscopic”.
73
Inclusion criteria included patients of any age undergoing a primary and revision
74
elbow arthroscopy, which may have been performed in conjunction with other
75
concomitant procedures. Exclusion criteria included incorrectly coded procedures
76
where arthroscopy was not performed and no postoperative follow-up. All operative
77
notes, orthopedic outpatient and inpatient visit documentation, and urgent care and
78
emergency room visits were reviewed. All patients were part of an integrated single
79
payer health system with a single electronic medical record. One patient had no further
80
encounters after his initial surgery and was removed from the analysis of postoperative
81
complications. Release from medical care was defined as the length of time from
82
surgery to the last visit with an orthopedic provider.
83
Patient demographic information was recorded, as well as smoking status, body
84
mass index, medical comorbidities, hand dominance, surgical history of operative
85
extremity, preoperative, intraoperative and final postoperative elbow range of motion
86
of the operative extremity, and need for additional procedures on the operative
87
extremity. Due to limitations in charting and variability amongst providers, any patient
88
with a listing of diabetes, prediabetes, or high blood sugar on his medical record
89
problem list was categorized under a single “elevated blood sugar” category.
90
Surgical variables such as preoperative diagnosis, detailed nature of procedures
91
performed, tourniquet time, patient position, portals used, need for concomitant ulnar
92
nerve procedure, need for unplanned arthrotomy, use of local anesthesia, and use of
93
intraarticular steroid were recorded. Complications such as superficial infection, deep
94
infection, nerve injury, vascular injury, compartment syndrome, heterotopic ossification,
95
future surgeries, deep vein thrombosis/ pulmonary embolism, and re-hospitalization
96
were also documented. Deep infection was defined as any postoperative infection
97
requiring hospitalization, intravenous antibiotics and/or surgical debridement.
98
Superficial infection was defined as any patient prescribed post-operative oral
99
antibiotics which did not require hospitalization or surgical irrigation and debridement.
100
Nerve damage was further stratified by nerve injured: ulnar, radial, median or medial
101
antebrachial cutaneous nerve. Any patient with concomitant open ulnar nerve
102
procedure was excluded from the analysis of ulnar nerve damage, as it would be unclear
103
if the nerve deficit were related to the arthroscopy or ulnar nerve procedure.
104
Surgeon characteristics were gathered including fellowship training, years in
105
practice, and number of cases performed during study period. For the purpose of this
106
study, pediatric patients were defined as age eighteen and younger at time of the
107
surgery. One of the authors who did not perform any of the surgical procedures
108
compiled all the data through a retrospective chart review.
109
To aid in the analysis of risk factors and the perceived risk of the procedure, the
110
complexity scale designed by Nelson et al was modified (Table 2).1 This is not a validated
111
scale. Nelson et al tracked the factors contributing to the complexity rating (ranging
112
from 1 to 9) including type of procedure (scored as 1 to 5), tourniquet time (scored as 0-
113
2), and number of portals used (scored as 0-2). The procedure was then categorized as
114
low, medium, or high complexity based on its numeric score. In the modified complexity
115
scale, the type of procedures scored was expanded to include a wider variety of
116
procedures. Nelson et al allotted a bonus point for release of posterior band of medial
117
collateral ligament or medial epicondylectomy. In the modified scale, the bonus point
118
was awarded for any concomitant ulnar nerve procedure.
119
Statistical Analysis
120
Statistical calculations were performed using a Binary Logistic Regression
121
analysis to fit a logistic regression model to investigate the relationship between
122
discrete outcomes with binary levels and a set of categorical predictors. Results were
123
determined to be statistically significant with a p–value of less than 0.05.
124 125 126
RESULTS Five hundred and seventy-three procedures were identified. Of those, 560
127
elbow arthroscopies in 528 patients met our inclusion criteria. Twelve procedures were
128
excluded due to incorrect inter-facility code. One procedure was excluded due to lack of
129
followup. Of those procedures, 447 (79.8%) were performed in adults and 113 (20.2%)
130
in pediatric patients. The mean age at the time of surgery was 38.6 years (range 5 – 88
131
years). The mean age in adults was 44.5 years (19-88 years) and the mean age in
132
pediatric patients was 15.2 years (5-18 years). 79.1% of the procedures were performed
133
in male patients. Five hundred and twelve patients (91.4%) did not use tobacco at the
134
time of the procedure. Fifty procedures (8.9%) were performed in patients who had
135
diabetes and twenty patients (3.6%) had prediabetes at the time of their procedure.
136
Thirteen patients (2.3%) had a diagnosis of rheumatoid arthritis and two patients (0.4%)
137
had a diagnosis of juvenile rheumatoid arthritis at the time of their procedure. One
138
hundred and nine patients (19.5%) had had a previous elbow surgery on the operative
139
extremity. Twelve procedures (2.1%) received and intra-articular steroid injection. Pre-
140
procedure diagnosis included elbow arthritis, symptomatic loose bodies, elbow
141
contracture, valgus extension overload, impingement, instability, osteochondritis
142
dessicans, lateral epicondylitis, septic joint, and elbow pain. The complexity of cases is
143
summarized in Figure 1.
144
The average length of follow-up was 376 days (range 2- 2739 days). Five patients
145
had less than a week follow-up. An additional forty-three patients had less than two
146
weeks follow-up. Fifty-nine elbows (10.5%) required a future surgery on the operative
147
elbow (Table 3). Seventy-one patients (12.7%) had a concomitant ulnar nerve
148
procedure. These were excluded from the analysis of ulnar nerve complications
149
postoperatively.. 367 (65.5%) procedures were performed in the prone position, 178
150
(31.8%) in lateral, and 10 (1.8%) in supine and 1 (0.2%) in beach chair.
151
Of the 42 board certified orthopaedic surgeons, 33 (78.6%) of surgeons were
152
fellowship trained. The fellowships were in sports medicine (38.0%), hand/wrist
153
(31.0%), shoulder and elbow (7.1%) or upper extremity (2.4%). Of the operating
154
surgeons, 33 (78.6%) performed less than 10 elbow arthroscopies in the study sample, 5
155
(11.9%) performed 11-50 elbow arthroscopies, 3 (7.1%) performed 51-100 and 1 (2.4%)
156
performed greater than 100 elbow arthroscopies in the study period. Ninety-two
157
(16.4%) procedures were performed during the surgeon’s first 5 years in practice. Two
158
hundred and forty-one (43.0%) procedures were performed between the surgeon’s 6
159
and 10 years in practice. Ninety-five (17.0%) procedures were performed between the
160
surgeon’s 11 and 15 years in practice. Forty-one (7.7%) procedures were performed
161
between the surgeon’s 16 and 20 years in practice. Sixteen (2.9%) procedures were
162
performed after the surgeon had been in practice more than 20 years. Surgeon based
163
data was unavailable for seventy-five procedures.
164 165
Complications
166
Nerve Injuries
167
The most frequent reported complication was transient neurologic
168
complications. Of the 560 procedures, there were twenty transient nerve palsies (3.5%).
169
Nerve injuries included: 8 (1.4%) ulnar (7 sensory, 1 mixed motor/sensory), 8 (1.4%)
170
radial/posterior interosseous nerve (6 sensory, 1 motor to posterior interosseous nerve,
171
1 mixed motor/sensory radial nerve), 3 (0.5%) medial antebrachial cutaneous nerve, and
172
1 (0.2%) median nerve (sensory). Table 4 summarizes the details of each case. Previous
173
elbow surgery (OR 3.23, P=0.004) and female sex (OR2.96; P=0.008) had a significantly
174
higher risk of nerve injury (Table 5).
175
176 177
Infection There were 3 (0.5%) deep infections and eleven (2%) superficial infections. One
178
deep infection was culture negative, one had positive mycobacterium avium cultures
179
which eventually required resection arthroplasty. A third late infection (6 month post
180
op) was culture positive for streptococcus viridans. Table 6 summarizes all the variables
181
that may play a role in causing an infection. The only variable that reached significance
182
for infection was elevated blood sugars. Patients with elevated blood sugars had a 4.11
183
odds (95% CI: 1.337, 12.645; P=0.0136) of developing a post-operative infection. There
184
were no infections in patients who were smokers or those who had a steroid injection at
185
completion of case.
186
187 188 189
Heterotopic Ossification
190
with nine (1.6%) requiring additional surgeries for resection. All cases of HO occurred in
191
male patients, but no other significant risk factors were identified (Table 7), Figure 2.
192
Adult vs Pediatric
Fourteen (2.5%) elbows developed postoperative heterotopic ossification (HO)
Relative to pediatric patients, there no statistically significant increased risk
193 194
adults for nerve injury (OR 1.99; P=0.27), infection (OR 3.286; P=0.24), and heterotopic
195
ossification (OR 1.40; P=0.66). There was no difference in re-operation rate (OR 0.92,
196
P=0.8).
197 198 199
Other Complications There were no cases of vascular injury, compartment syndrome, deep vein
200
thrombosis, or pulmonary embolism. Three patients required re-hospitalization for a
201
deep infection post operatively.
DISCUSSION
202 203
In this large case series with forty-two different operating surgeons, 3.5% of
204
procedures developed a transient nerve palsy. Previous elbow surgery (OR 3.57,
205
P=0.006) and female sex (OR4.05; P=0.002) were significant risk factors for nerve injury.
206
In addition, 2.5% of cases developed postoperative heterotopic ossification, all of whom
207
were male. No other significant risk factors were identified. Relative to pediatric
208
patients, there was a higher odds in adults for nerve injury, infection, and heterotopic
209
ossification, but none reached significance. We also found that patient factors (BMI,
210
age), surgical factors (patient position, tourniquet time, complexity) and surgeon factors
211
(fellowship training, case volume, years of practice) did not have a significance in
212
neurologic complications.
213
Reported nerve injury rates range from 0.5% -10.4% (Table 7). 1-8 Recent large
214
case series report neurological complications of approximately 2%. These are lower than
215
our reported 3.5% nerve injury rate. The reason behind this discrepancy is unclear. One
216
potential explanation is our inclusion of concomitant arthroscopic and open procedures.
217
We chose to include all surgeries with an arthroscopic component as we felt this was
218
more indicative of everyday community practice.10 ,11 Many prior papers have focused
219
on arthroscopic surgeries alone. Four hundred and twenty one (75%) of our procedures
220
were exclusively arthroscopic.
221
Heterotopic ossification may be an underreported complication of elbow
222
arthroscopy. In this series, 2.5% of cases developed postoperative heterotopic
223
ossification with 1.6% of cases requiring repeat surgical procedures as a result of the
224
heterotopic ossification. All cases occurred in male patients, but no other significant risk
225
factors were identified. Heterotopic ossification prophylaxis was not routinely utilized.
226
Similar to our results, Nelson et al reported 6 cases (1.4%) of heterotopic ossification
227
requiring future surgeries. Nelson notes that postoperative XR were not regularly
228
obtained and that the six-week follow-up of his study may have led to the under
229
reporting of this complication. Likewise, not all surgeons in our study obtained routine
230
postoperative XR. Future studies are warranted to further examine the risk of post-
231
operative heterotopic ossification and possible prevention strategies.
232
Post- operative infection is an important complication of elbow arthroscopy. We
233
found a 2% superficial infection and 0.5% deep infection which was similar to published
234
results.12 Nelson et al published a substantially higher infection rate (6.7% superficial
235
infection and 2.2% deep infection). Previous studies have documented the use of intra-
236
articular steroid injection as a risk factor for infection. In this study, steroid injection was
237
not commonly performed (2.1% of procedures). No infections were seen in patients
238
who received a steroid injection. Overall, elevated blood sugar had a significantly higher
239
risk for infection (OR 4.11, P=0.0136). Smoking status, body mass index greater than 40,
240
tourniquet time and previous elbow surgery did not significantly increase the risk of
241
infection. There were no cases of vascular injury, compartment syndrome, deep vein
242
thrombosis, or pulmonary embolism.
243
Our study is unique in that it contains a large pediatric cohort. Few previous
244
large-scale studies have been published on elbow arthroscopy in the pediatric
245
population.13 Vavken et al published on the experience of the first 50 elbow
246
arthroscopies performed in a pediatric orthopedic department in Switzerland. They had
247
3 transient nerve palsies (6%) in this series.14 Micheli et al also published on their
248
experience with 47 elbow arthroscopies in the pediatric and adolescent population.15
249
They had no nerve palsies in their series. We found two nerve palsies out of the 114
250
pediatric elbow arthroscopies studied (1.8%) – of these palsies, one was the ulnar nerve
251
and one was the radial nerve. Relative to pediatric patients, there was a higher risk in
252
adults for nerve injury (OR 1.99; P=0.27), infection (OR 3.286; P=0.24), and heterotopic
253
ossification (OR 1.40; P=0.66). However, none of these trends reached statistical
254
significance. There was no difference in re-operation rate (OR 0.92, P=0.8).
255
Strengths
256 257
The strengths of this study includes the large number of cases and contributing surgeons which help provide a realistic analysis of post-operative complications that can
258
be applicable to many community orthopedic surgeons. The inclusion of surgeries with
259
both an arthroscopic and concomitant open component increases the applicability of
260
these results and aides with preoperative patient counseling. This study examined
261
specific risk factors for postoperative complications and determined odds ratios for each
262
risk factor identified. This allows for a more in-depth analysis than database studies
263
allow.
264
Limitations
265
The weakness of this study is that it is retrospective. The integrated nature of the health
266
system allows for the vast majority of patients to be captured in a single electronic record.
267
However, a patient could theoretically seek care outside of the health care system for a
268
postoperative complication. This would be unlikely given the financial implications of an
269
outside facility. In addition, the small number of complications made statistical analysis
270
difficult and the study was not powered to reflect differences between the adult and
271
pediatric populations. The patients with concomitant ulnar nerve releases were excluded
272
from the analysis of post-operative ulnar nerve complications. Given that these patients
273
complained of pre-operative ulnar nerve symptoms, the authors were unable to tell from
274
the medical record if post-operative symptoms were a result of the underlying nerve
275
health or a result of a new injury. This may lead to an under reporting of ulnar nerve
276
injuries. In addition, the complexity scale we used was not validated. In our series, 48
277
patients had less than 2-week follow up which may have an inadequate length of follow-
278
up to properly analyze certain complications. Additional limitations include no defined
279
minimum follow up period, postoperative elbow stiffness and range of motion were not
280
analyzed complications in this study.
281 282 283 284
Conclusions Elbow arthroscopy is a safe procedure with low complication rates. Diabetes is a risk factor for infection. Prior surgery and female sex are at higher risk for nerve injury.
285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
References 1. Nelson GN, Wu T, Galatz LM, Yamaguchi K, Keener JD. Elbow arthroscopy: early complications and associated risk factors. JSES 2014(23):273-278. 2. Andrews JR, Carson WG. Arthroscopy of the elbow. Arthroscopy 1985(1):97-107. 3. O’Driscoll SW, Morrey BF. Arthroscopy of the elbow: Diagnostic and therapeutic benefits and hazards. J Bone Joint Surg Am. 1992(74): 84-94. 4. Reddy AS, Kvitne RS, Yocum LA, Elattrache NS, Glousman RE, Jobe FW. Arthroscopy of the elbow: a long-term clinical review. Arthroscopy 2000 (16): 588-94. 5. Kelly EW, Morrey BF, O’Driscoll SW. Complications of Elbow Arthroscopy. J Bone Joint Surg Am. 2001 (83):25-34. 6. Jinnah AH et al. Peripheral Nerve Injury After Elbow Arthroscopy: An Analysis of Risk Factors. Arthroscopy. 2018; 34(5): 1447-1452. 7. Marti D, Spross C, Jost B. The first 100 elbow arthroscopies of one surgeon: analysis of complications. JSES 2013 (22): 567-573. 8. Elfeddali R, Schreuder MH, Eygendaal D. Arthroscopic elbow surgery, is it safe? JSES 2013 (22): 647-652. 9. Schneider T, Hoffstetter I, Fink B, Jerosch J. Long-term results of elbow arthroscopy in 67 patients. Acta Orthop Belg. 1994 (60): 378-83. 10. Hilgersom NFJ, van Deurzen DFP, Gerritsma CLE, van der Heide HJL, Malessy MJA, Eygendaal D, van den Bekerom MPJ. Nerve injuries do occur in elbow arthroscopy. Knee Surg Sports Traumatol Arthrosc 2018: 26(1): 318-324. 11. Desai MJ, Mithani SK, Lodha SJ, Richard MJ, Leversedge FJ, Ruch DS. Major Peripheral Nerve Injuries After Elbow Arthroscopy. Arthroscopy. 2016; 32(6):999-1002. 12. Campl CL, Cancienne JM, Degen RM, Dines JS, Altchek DW, Werner BC. “Factors that Increase the Risk of Infection After Elbow Arthroscopy: Analysis of Patient Demographics, Medical Comorbidities, and Steroid Injections in 2,704 Medicare Patients. Arthroscopy. 2017: 33(6); 1175-1179. 13. Andelman SM, Meier KM, Walsh AL, Hausman MR. Pediatric elbow arthroscopy: indications and safety. J Shoulder Elbow Surg. 2017: 26(10):1862-1866. 14. Vavken, Muller, Camathias. First 50 Pediatric and Adolescent Elbow Arthroscopies: Analysis of Indications and Complications. Journal of Pediatric Orthopaedics. 2016: 36(4): 400-404. 15. Micheli, Luke, Mintzer, Waters. Elbow Arthroscopy in the pediatric and adolescent population. Arthroscopy. 2001: 17(7):694-9.
323 324
Table 1: Published Nerve Injuries after Elbow Arthroscopy Paper
Year
Number
Total nerve
Ulnar
Radial
MABC
Median
Other
Jinnah et al
Nelson et al Marti et al
injuries (%)
2018
253
15 (5.9%)
11
0
1
1
2
2014
417
7 (1.7%)
5
0
0
1
1
2013
100
2 (2%)
1
0
0
1
0
2013
200
4 (2%)
3
1
0
0
0
2001
473
12 (2.5%)
5 (1 transected)
5
1
1
0
2000
172
1 (0.5%)
1 (transected)
0
0
0
0
8
1994
67
7 (10.4%)
2
4
0
1
0
3
1992
71
3 (4%)
0
3
0
0
0
1985
12
1 (8.3%)
0
0
1
0
0
1
6
Elfeddali et al. Kelly et al
of Cases 6
7
5
Reddy et al
4
Schneider et al O’Driscoll et al Andrews et al
325
2
326
Table 2: Modified Elbow Arthroscopy1 Complexity Scale Modified Elbow Arthroscopy Complexity Scale Procedure Limited debridement (1 compartment), diagnostic arthroscopy, irrigation and debridement Extensive debridement (2+ compartments), loose body removal, plica excision, osteophyte/mass excision, olecranon bursectomy Capsular release, microfracture, radial head resection, ligament repair, lateral epicondylar release Capsular resection with resection of bone, osteochondral allograft, osteochondral lesion fixation, Outerbridge-Kashiwagi, radial head replacement/ ORIF, UCL reconstruction Ulnar Nerve Procedure Tourniquet Time ≤ 60 minutes 60-90 minutes > 90 minutes Portals ≤2 3-4 >4 Complexity Category Low Moderate High
Points 1 2 3 4
+1 0 1 2 0 1 2 0-3 4-5 >5
327 328
Table 3: Repeat Operative Procedures after Elbow Arthroscopy Repeat Operative Procedures Ulnar nerve decompression
11
Open irrigation and debridement, loose body removal
10
Repeat arthroscopic debridement
9
Heterotopic ossification resection
8
Total elbow arthroplasty
5
Open contracture release
5
Radial nerve decompression
4
Radial head replacement
3
Open elbow exploration
2
Medial collateral ligament reconstruction
1
Osteocapsular arthroplasty
1
Scar excision
1
Osteochondral defect surgery
1
Medial epicondylar release
1
Lateral epicondylar release
1
329 330
Table 4: Nerve Injuries after Elbow Arthroscopy Patient ID
Nerve Injured
Procedure Performed
61F
Radial nerve sensory palsy improved at 4 months
Capsular release and osteophyte removal
43F
Radial nerve sensory palsy improved at 4 weeks
Arthroscopic extensor carpi radialis brevis debridement
Low
15M
Radial nerve sensory palsy resolved at 3 months
Anterior capsular release
Moderate
33M
Radial nerve sensory palsy resolved by 6 weeks
Debridement of plica, ulnar nerve decompression
Moderate
50F
Radial nerve sensory palsy resolved by 7 months
Arthroscopic debridement, open capitellar resurfacing using tendon allograft, ulnar nerve decompression
Moderate
36M workers compensation
Radial nerve sensory palsy improved at 11 months
Debridement, capsulectomy, synovectomy, osteophyte resection, ulnar nerve transposition
High
57F
Posterior interosseous nerve motor palsy improving at 6 months
Debridement, capsulectomy, partial synovectomy
Moderate
35F with rheumatoid arthritis and diabetes mellitus
Radial nerve motor and sensory palsy with post operative septic elbow, nerve palsy resolved by 2 years
Debridement followed by arthroscopic irrigation and debridement
Moderate
20F
Median nerve sensory palsy resolved by 2 months
Posteromedial elbow debridement, microfracture
Moderate
Modified Complexity Level Low
331 332
45M with HTN, hyperlipidemia and diabetes
Ulnar nerve sensory palsy resolved by 2.5 months
Debridement of plica
Low
53F
Ulnar nerve sensory palsy continued at 2 weeks
Debridement, capsular division, partial synovectomy
Moderate
17F
Ulnar nerve sensory palsy resolved by two years
Arthroscopic posterior osteophyte excision and release with open anterior release
Moderate
31M
Ulnar nerve sensory palsy resolved by 2 months
Debridement and medial collateral ligament reconstruction using allograft tissue
Moderate
60M with HTN, CAD, Ehlers Danlos, emphysema
Ulnar nerve sensory palsy resolved by 6 weeks
Outerbridge-Kashiwagi procedure
Moderate
28M
Ulnar nerve sensory palsy resolved by 6 weeks
Osteocapsular arthroplasty
High
52M
Ulnar nerve sensory palsy, improved 4m post op
Synovectomy
Moderate
38M
Ulnar nerve sensory and motor palsy, underwent ulnar nerve exploration and release at 2 years with intact nerve
Debridement, osteophyte resection, capsular division, partial synovectomy, injection of amniotic cells into joint space
High
67F with hypertension, hyperlipidemia and obesity
Medial antebrachial cutaneous nerve sensory palsy improving at 6 weeks
Endoscopic carpal tunnel release, open anterior ulnar transposition, elbow arthroscopy, removal of loose body and debridement
Moderate
24F with history of alcohol abuse and anxiety
Medial antebrachial cutaneous nerve sensory palsy improving at 3 months
Elbow arthroscopy, debridement, osteophyte resection, capsular division, synovectomy, loose body removal, ulnar nerve transposition
High
64M
Medial antebrachial cutaneous nerve sensory palsy improving at 2 weeks, resolved by 2 years
Elbow arthroscopy and extensive debridement, removal of loose body
High
333
Table 5 summarizes the odds of each variable for sustaining a nerve injury Odds ratio
95% CI
P-value
Age >18 years
2.328
0.532
10.183
0.2616
BMI >40
3.625
0.775
16.965
0.1019
Tourniquet Time >60 minutes
0.983
0.350
2.762
0.9735
Previous Elbow Surgery
3.567
1.440
8.839
0.0060
Moderate Complexity (4-5)
2.921
0.812
10.513
0.1009
High Complexity >5
1.985
0.466
8.453
0.3534
Supine vs lateral positioning
2.714
0.301
24.494
0.3737
Prone vs lateral positioning
0.826
0.319
2.135
0.6928
Supine vs prone positioning
3.287
0.385
28.067
0.2768
Female sex
4.047
1.642
9.970
0.0024
More than 5 years post training
2.397
0.312
18.397
0.4004
Smoker (yes)
0.552
0.072
4.216
0.5668
Fellowship-trained (Yes)
0.722
0.092
5.679
0.7567
334 335
Table 6 summarizes the odds of each variable for developing an infection Odds ratio
95% CI
P-value
Age >65 years
1.090
0.139
8.566
0.9349
Age >18 years
3.355
0.434
25.918
0.2459
BMI >40
2.385
0.295
19.289
0.4152
NA
NA
NA
0.9739
Elevated blood sugar
4.112
1.337
12.645
0.0136
Tourniquet Time > 60 minutes
0.527
0.169
1.639
0.2686
History of previous elbow surgery
2.340
0.768
7.128
0.1346
Moderate Complexity (4-5)
1.061
0.295
3.820
0.9273
High Complexity (>5)
1.173
0.289
4.788
0.8209
More than 4 portals
0.488
0.063
3.782
0.4919
Male
1.601
0.353
7.254
0.5416
Smoker (yes)
More than 5 years in practice
0.801
0.173
3.701
0.7761 0.5158
Fellowship-trained
336 337
Table 7 summarizes the odds of each variable for developing heterotopic ossification Odds ratio
95% CI
P-value
Age >18 years
1.403
0.307
6.417
0.6627
BMI > 40
2.583
0.317
21.020
0.3749
Tourniquet Time > 90minutes
1.910
0.529
6.903
0.3234
History of previous elbow surgery
1.237
0.335
4.573
0.7499
Moderate Complexity (4-5)
2.157
0.430
10.818
0.3500
High Complexity >5
2.996
0.573
15.677
0.1938
Use of small arthroscope
1.813
0.226
14.517
0.5753
Smoker (yes)
0.906
0.115
7.123
0.9251
Fellowship-trained
0.5312
338 339 340
Figure 1: Number of cases performed in each complexity category as detailed in the Modified Elbow Complexity Scale (Table 2)
341 342 343 344
Figure 2: Representative case of 40 year old male with left elbow contracture (A) Patient underwent arthroscopic osteocapsular arthroplasty. (B) Initial post-operative radiographs. (C) Patient subsequently developed heterotopic ossification requiring open heterotopic ossification excision.